Hello,

I used to get very similar failures (a while ago) with NetBSD + alpha (a 
miata) + qlogic 1040.  My solution (lame) was to go to ebay and pick up 
a symbios card (siop driver).  Have you suddenly increased disk 
activity?  Also, as per the other replies, try a newer version.  FWIW 
NetBSD 1.6beta rocks.

skd

Chris Rupnik wrote:

> Oh well,
>  Happened again last night. The first time it happened around 7:20am, this
> time at 1:10am. Nothing running at the time.
> Any ideas? I'm planning a 1.6 upgrade anyhow. So, if that is the fix, so be
> it :)
>
> Chris
>
> isp0: Mailbox Command 'ABORT' failed (TIMEOUT)
>isp0: Polled Mailbox Command (0x15) Timeout
>isp0: Mailbox Command 'ABORT' failed (TIMEOUT)
>raid0: IO Error. Marking /dev/sd3a as failed.
>raid0: node (Rod) returned fail, rolling backward
>raid0: DAG failure: w addr 0x17b0a0 (1552544) nblk 0x10 (16) buf
> 0xfffffe00011f2000
>isp0: Polled Mailbox Command (0x15) Timeout
>isp0: Mailbox Command 'ABORT' failed (TIMEOUT)
>raid0: IO Error. Marking /dev/sd1a as failed.
>raid0: node (Rod) returned fail, rolling backward
>isp0: Polled Mailbox Command (0x15) Timeout
>isp0: Mailbox Command 'ABORT' failed (TIMEOUT)
>raid0: node (Rod) returned fail, rolling backward
>raid0: DAG failure: w addr 0x17b090 (1552528) nblk 0x2 (2) buf
> 0xfffffe0001812000
>Multiple disks failed in a single group! Aborting I/O operation.
>Multiple disks failed in a single group! Aborting I/O operation.
>Multiple disks failed in a single group! Aborting I/O operation.
>Multiple disks failed in a single group! Aborting I/O operation.
>[Failed to create a DAG]
>panic: raidframe error at line 455 file
> ../../../../dev/raidframe/rf_states.c
>Stopped in raid at cpu_Debugger+0x4: ret zero,(ra)
>db> trace
>cpu_Debugger() at cpu_Debugger+0x4
>panic() at panic+0xfc
>rf_State_CreateDAG() at rf_State_CreateDAG+0x224
>rf_ContinueRaidAccess() at rf_ContinueRaidAccess+0xdc
>rf_ContinueDagAccess() at rf_ContinueDagAccess+0x22c
>DAGExecutionThread() at DAGExecutionThread+0x1cc
>esigcode() at esigcode
>--- root of call graph ---
>db>
>
>
>
>